Fixed Platform Performance During Recent Hurricanes: Comparison to Design Standards

Abstract

Abstract Hurricanes Ivan, Katrina and Rita passed through over 3,000 platforms in the Gulf of Mexico during 2004 and 2005. While most structures performed adequately, 123 were destroyed and over 183 had major above and or below water damage. There was no life loss and no significant pollution, which is a tribute to the Minerals Management Service (MMS) oversight and American Petroleum Institute (API) design codes, specifically API RP2A (RP2A) for the structural design of fixed offshore platforms. RP2A has evolved considerably over the past 38 years since its first edition in October 1969. Some of this is attributable to changes in engineering practice and the results of experimental studies. Some is attributable to the experiences and lessons learned in large storms and hurricanes. These three recent hurricanes resulted in the largest number of destroyed and damaged platforms in the history of Gulf of Mexico operations. There are several industry, API and MMS studies underway to understand what went right and what went wrong, and to update industry guidelines and practices accordingly. This paper presents the initial results of a study performed for the MMS on the performance of fixed platforms in these hurricanes. The paper also ties-in findings of the work and its correlation to RP2A design practices. Background While hurricanes are not a welcome sight in the Gulf of Mexico, they do provide opportunity to observe how platforms perform in the storms and if required, an opportunity to update design codes accordingly. In fact, development of RP2A was in part driven by several hurricanes in the late to mid 1960's including Carla (1961), Hilda (1964), Betsy (1965) and in particular hurricane Camille (1969), which was one of the largest hurricanes to impact platforms in the Gulf of Mexico. All of these hurricanes damaged or destroyed platforms and the industry responded with the development of RP2A. In the 1970's, RP2A further evolved and the platforms were tested by hurricanes Carmen (1974) and Frederic (1979). The 9th edition of RP2A was issued in November 1977 and contained the first industry accepted wave load "recipe" including the use of 100 year return period conditions and consistent hydrodynamic drag and inertia coefficients. The 9th edition represents a substantial improvement in platform design, and is in fact demonstrated by the better performance of 9th edition and later RP2A designs in these hurricanes, as describe later. In 1985 hurricane Juan spawned in the Gulf of Mexico as a "Sudden Hurricane" and the industry was not able to evacuate all of the platforms. Several manned platforms were severely damaged, including several that toppled during the storm. Fortunately, there was no direct life loss associated with the fixed platform failures. Later investigation determined that the platforms were older vintage and in need of repair prior to Juan. It was clear that specific guidelines were required for assessment of existing offshore platforms for fitness of purpose. The industry, including significant support from the MMS, began a series of joint industry and special projects, as well as API initiatives to develop an assessment standard for existing platforms. The result was RP2A Section 17, developed in the early 1990's and originally published in 1996 as a supplement to the 20th edition. Interestingly, in 1992 during the development of Section 17, hurricane Andrew damaged or destroyed dozens of older Gulf of Mexico platforms. A hurricane the size of Andrew had not occurred since Camille and provided an opportunity to both test and then calibrate Section 17 during its development. In 2002 hurricane Lili damaged and destroyed several older platforms, something that had not been see